* Relu6 is the bottleneck op for Mobilenet-v2. As the mkldnn supports the conv/relu6 fusion, we implement it fusion via cpass way. Due to the int8 enabling for this fusion will be supported in MKLDNN v0.20, so this PR is focused on the fp32 optimization.

Below table shows the benchmark(FPS) which measured on skx-8180(28 cores)
Batch size | with fusion | without fusion
-- | -- | --
1 | 214.7 | 53.4
50 | 1219.727 | 137.280

test=develop

* Fix the format issue

test=develop

* Add the missing nolint comments.

test=develop

* Fix the typos.

test=develop

* Register the conv_brelu_mkldnn_fuse_pass for the MKLDNN engine.

test=develop

* Adjust the indentation.

test=develop

* Add the test_conv_brelu_mkldnn_fuse_pass case.

test=develop

* Slightly update the code per Baidu comments.
Let the parameter definition embedded into the code.
That's will make the code easy to understand.

test=develop

* Add LAMB optimizer

* Expose LAMB Optimizer's APIs

test=develop, test=document_preview

* Cleanup code & doc

test=develop, test=document_preview

* Update lamb optimizer's formula

test=develop

test=develop

http://newicafe.baidu.com:80/issue/PaddleSec-33/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-28/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-25/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-24/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-21/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-20/show?from=page

test=develop

* add quant_dequant_moving_avg_max_abs op
test=develop

* add more note for quantdequant op
test=develop

* optimize communicator flag

* change flags in init py test=develop

* double backward, elementwise_div

* fix dx empty. test=develop

* bug fix (#17392)

fix secure bug

* Eanble stack operator for a Ngraph, test=develop (#17406)

* fix sqrt_grad_grad unittest. test=develop (#17410)

* fix sqrt_grad_grad unittest. test=develop

* disable sqrt_grad_grad unittest. test=develop

* test=develop, fix unittest

* test=develop, fix unittest

* test=develop, fix unittest

* test=develop, fix bug

* fix unittest. test=develop

* fix unittest dx. test=develop

* tmp fix! for test... test=develop

* reduce tmp, test=develop

* test=develop, reduce tmp

* fix broadcast unittest. test=develop

* fix format. test=develop

* refine code. test=develop

* refine code. test=develop

* refine GetDoubleGradSafeTensor. test=develop

* fix format. test=develop

add elementwise_sub_grad_grad op for backward of backward calculation

* add record_event
test=develop

* remove csp
test=develop

* add var grad hook test=develop

* improve gru unit performance.
refine code

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* Add conditional compile for gru opt

Not enable gru opt if compute ability < 700

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* refine code.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

test=develop

* double backward sqrt

* refine unittest. test=develop

* refine test. test=develop

* remove alpha in unittest. test=develop

* test=develop, double backward reduce_mean

* add comment. test=develop

* fix format. test=develop

* rename GradGrad -> DoubleGrad. test=develop

* fix op_use_default_grad_op_maker.spec. test=develop

* add elementwise_add_grad_grad op. test=develop

* use defined GradMaker. test=develop

* add collect fpn proposals op,test=develop

* add double grad for mul_op. test=develop

* fix format. test=develop

* fix format. test=develop

* fix format. test=develop

* refine code. test=develop

* remove setzero. test=develop

* fix dx/dy init bug. test=develop

* fix format. test=develop

test=develop

* Optimize the computing kernel of sequence_reverse operator.

test=develop

* Clean code

test=develop

* Fix for cpplint syntax checking.

test=develop

* Fix the compile warning issue.

test=develop

* Optimize the elementwise op with CUDA kernels.
test=develop

* Support setting of attr in op config file.
test=develop

* Add the support the setting dtype and initializer in config.
test=develop

* Save workspace.

* Add initializer "zeros".
test=develop

* Fix compiling error.

* Support the use of existed file to initailize tensor in op_tester.

* Use eigen to optimize the elementwise_add/mul for the case that x and y have the same dims.
test=develop

* add double grad for elementwise_mul. test=develop

* remove comment. test=develop

* fix grad sum. test=develop

* fix for axis expand. test=develop

* add test for axis expand. test=develop

* add double grad for square. test=develop

* formax code. test=develop

* fix for grad sum. test=develop

* refine shape. test=develop

* refine extract. test=develop

refine code

fuse cublas  calling and kernels into one cuda kernel.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* Add conv2d_grad_grad_op
* Extracte the cuDNN conv algo searching code in conv_cudnn_helper.h.
    - Now use it in conv2d_grad_grad.
    - Will simply the searching code in conv2d and conv2d_grad in next PR.
* Enhance and fix bug in unit testing of gradient_checker.
* Support to fetch empty variables，return None in Python.

* test=develop
add elementwise_mod and elementwise_floordiv, fix equation problem in elementwise_mod

* modified formula for lrn

test=develop

* modified api.spec

test=develop

* Refine elementwise kernel.

Add a simple cuda kernel if grad x and y both exist
Use 2D block cuda kernel to do broadcast.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* refine code.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* refine code.

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>